Abstract: A closed-loop PD-type iterative learning control scheme for nonlinear systems is proposed in this paper,and sufficient and necesary conditions for the convergence of the proposed learning scheme are given.The convergence of the proposed learning scheme is just related to D-type learning gain,and is independent of P-type learning gain and the idiographic form of state functions.

Key words: iterative learning； convergence； nonlinear system； closed-loop PD-type；